step 1: what’s the point?

Predicting the future.

We seek to better understand what’s happening on the court, to understand this sport better, and learn some truth about it – and the way this should manifest itself is: being better at knowing what’s going to happen than average.

step 2: point score %

PS% has some magic to it. The team with the higher PS% in the set is guaranteed to win. This is easy then! We’re done!

Unfortunately, it may not be that simple. This is the problem Nate Silver of FiveThirtyEight describes as the signal & the noise. PS% data is noisy, a lot of factors contribute to this single number.

step 3: attack efficiency

So we know that PS% is noisy – this makes using it to predict the future rather tricky. To that extent, we need something more granular.

We start with attacking because it’s a single skill & has the highest direct impact on points scored. So, can we predict the future if we knew the attack efficiencies of the two teams leading up to a match?

step 4: passer ratings

I think passer ratings are silly.

And while this might be a stupid hill to die on, it’s pretty obvious that passer ratings do not adequately represent reality.

In a common 3pt scale, 3-2-1-0, there’s an assumption that the difference between each of the 4 outcomes is identical. The value of going from 2 to 3 is the same as going from 0 to 1 for example.

But you tell me, two kids both pass a 1.5 on 10 balls. Which one do you want?

Player A: 2 – 2 – 2 – 2 – 2 – 1 – 1 – 1 – 1 – 1

Player B: 3 – 3 – 3 – 3 – 3 – 0 – 0 – 0 – 0 – 0

step 5: if he’s a good passer, why doesn’t he pass good?

Are we using the right tools to measure performance?

step 6: if he’s a good hitter, why doesn’t he hit good?

But does he get on base?

step 7: we shall see

step 8: we shall see also…

step n-1: Expected Value (EV) or Expected Points (xP) or something…

step n: predicting the future